Interface: TrustedContentAttrs
Defined in: src/batteries/llm/chat_common/types.ts:125
Attribute bag rendered onto the trust envelope that wraps trusted (first-party) content.
Properties
| Property | Type | Description | Defined in |
|---|---|---|---|
kind | string | Content classifier rendered on the envelope. | src/batteries/llm/chat_common/types.ts:137 |
modality? | "opaque" | "inert" | "extractable" | Same semantics as UntrustedContentAttrs.modality. | src/batteries/llm/chat_common/types.ts:143 |
nonce | string | Nonce binding the envelope's open and close markers together — it becomes part of the tag NAME (<..._<nonce>>) so a model mirroring the fence cannot forge a sibling's closer. FOOTGUN: this nonce MUST be unguessable AND non-path-shaped. It is typically the source record's id; a path-shaped id (e.g. chunk-assembly-events-9) is copied by small models as a CITATION (/assembly/events-9), which a doc-path validator then rejects → re-cite loop. Mint record ids with crypto.randomUUID(); carry page/human provenance in source= (rendered before nonce=), never in the id. | src/batteries/llm/chat_common/types.ts:135 |
tool? | string | Name of the tool that produced the content, when applicable. | src/batteries/llm/chat_common/types.ts:139 |